Fast Visual Saliency Map Extraction From Digital Video

A fast algorithm to extract the visual saliency map from digital video is proposed in this paper. Since, digital video is stored and/or transmitted in coded form, the authors utilize the Motion Vector Field (MVF) for foreground/background separation and apply an image based contrast model for further performance improvement. The performance of the proposed algorithm is demonstrated by experimental results. One important property of the Human Visual System (HVS) is visual attention, which refers to a relatively small region of an image sequence that attracts human attention the most. Locations of visual fixation in video define the visual saliency map.